Power Unit on a Wet Formation Lap: A Data Investigation into the Sepang Software Failure

The track at Sepang International Circuit was slick with rain. A few minutes remained before the Grand Prix. The formation lap had begun, the cars were out of the garages, and the radio channels were carrying one complaint after another — the car was moving, but it was not moving. The power unit was losing power. On the rain-soaked asphalt, cars came almost to a standstill, and the headline of an entire race weekend became a single line of software.

Those standing beside the track saw an uncomfortable image dissolve into the raindrops: a 'new era' car, meant to be the emblem of technical excellence, apparently forgetting its own language on the formation lap itself. Drivers such as Max Verstappen and Lewis Hamilton did not all treat the matter lightly — some compared it to battery-powered cars made of Lego. Once that sentence spread across social media, what followed was not technical analysis; it was a storm of public opinion.

I do not cover Formula One week to week. My work is in football data, where pass networks, pressing intensity and expected goals are used to reconstruct the truth of a match. Even so, the Sepang incident stopped me, because the structure of the failure that emerged — a condition-specific, previously untested fault, corrected late through a power-deployment decision — is one I recognise. Based on my years of watching matches, what I have learned is that the real story is never in the scoreline; it lives in the gap between conditions and the baseline.

Context: the 'new era' power unit, the formation lap and the geography of Sepang

To understand this, one must first understand what the formation lap does. Before a race, drivers complete a lap to bring the car and tyres to operating temperature. Engine speed is generally low, the tyres have not yet gripped, and the driver's attention is on strategy rather than maximum performance. The first clue hides here: when a car runs slowly, with low grip, in wet conditions, its software and energy management system operate in a different environment — one in which it was never tested on a dry track.

After the 2026 ground-effect regulations, Formula One power units changed substantially. A modern hybrid unit combines an internal combustion engine with an electrical component, and a software layer governs how power is shared between them. This layer decides when energy is released from the battery, when it is recovered, and which mapping applies in which track sector. In other words, not only the machine but a decision made in code determines the car's speed.

The geography of Sepang International Circuit plays a large role in this equation. It was a long-standing fixture on the Formula One calendar, which it has not hosted since 2026. It is a track of long straights and slow corners, where heavy rain rapidly removes grip. In such an environment, the car's software should have been stable — and in reality it was not.

Driver dissatisfaction is not new. Complaints about the new-era power unit — energy management, weight balance, limits on driver control — had continued since the opening race of the season. This season has been described in places as a 'difficult first season.' The Sepang incident is therefore not isolated; it is the most visible episode of a long-running debate.

There is a further layer here — who controls what. The Fédération Internationale de l'Automobile (FIA) sets the technical regulations, but the fine-grained mapping of the power unit's software and energy management rests largely with the manufacturers. Nikolas Tombazis, the FIA's single-seater director, publicly explained the failure, and that choice is significant: rather than the institution's most senior figure, a named technical authority was placed at the front. This is a deliberate posture of transparency, a familiar pattern in crisis management.

Core analysis: a failure of four conditions meeting

Now to the substance. According to the original report, the failure arose from the unintended interaction of four variables: low engine speed, a low-grip wet track, energy-management modes set for wet weather, and the layout of specific circuit sectors. When these four coincide, the power unit suffers an 'unintended loss of power.'

The key point here is that the failure belongs not to any single component but to the combination of conditions. This is a familiar pattern in football analysis. A team may be excellent at high pressing, but if the opponent sits in a deep block and the pitch is wet, that pressing loses half its effect — the components are identical, the conditions change.

Why the fault was not caught in dry testing is the most important question. The FIA's explanation is clear: insufficient data was gathered in wet conditions during winter pre-season testing. In other words, the car was not tested in the environment where the problem would appear. Where test coverage has a gap, failure is inevitable — an old lesson from software engineering, equally true in sports governance.

Now to the remedy, because this is where the incident becomes most instructive. While the race was suspended because of rain, the FIA released an over-the-air software update that removed the energy-limit constraint in the affected sectors. In other words, the problem was not merely identified — a correction was applied mid-race, from outside the track.

That decision is itself a message: a configurable software layer governs power deployment, and it can be changed centrally at any time. This will now become not only a technical matter but a subject of future regulatory debate.

Honesty is required here. We have no telemetry data, no lap times, no measured speeds. What the original report contains is largely descriptive — cars slowed, lost power, and a correction followed. So I bind every causal claim in this piece to a specific confidence level: high confidence in the explanation of the underlying mechanism, medium confidence in the fear of future recurrence, and low confidence in the debate over software-layer control.

This is where I want to draw a comparison, while stating clearly that it is structural, not substantive. When stadiums went silent in 2026, home advantage slipped from 43.3% to 33.3%. That was a natural experiment: remove the variable of crowd presence, and see what happens to outcomes. The Sepang incident raises a similar question, though in technology rather than football: which variable, when removed, restores normal system behaviour, and which variable, when added, breaks it? The lesson of 2026 was that change cannot be explained without understanding the baseline. The lesson of Sepang is the same.

If I placed this incident in a model, I would take four inputs: track moisture, engine speed, energy-management mode, and sector layout. The baseline would be normal performance on a dry track. The deviation would appear only when all four variables simultaneously reached extreme values. The model's output might be a probabilistic statement — 'in this specific combination, the risk of power loss is greatest.' But we do not have the data to run that model. And this is exactly where an analyst's honesty is tested: without data, an estimate cannot be passed off as fact.

The original report carries a source credit — writer Alan Baldwin, editing by Pritha Sarkar. The source is an international news report that quotes both driver statements and the FIA's technical statement. In such reporting, driver emotion and institutional explanation sit side by side, and the analyst's job is to see them separately.

The contrarian angle: the crisis narrative versus the reality of the track

The first narrative to emerge is that of a 'new-era crisis.' But there is a clear distinction between correlation and causation here that needs to be marked. Disorder was visible at a wet race weekend — that is true. But is the cause uncontrolled technology, or a specific, limited software fault that was corrected during the race itself? If the second is true, the size of the crisis is much smaller.

The drivers' remarks must also be read carefully here. A line about cars made of Lego is newsworthy, not analytically valuable. It is an expression of mental pressure, not technical evidence. When a player or driver exaggerates in reaction, that is not a measure of the system's failure — it is a measure of the state of the relationship. The friction visible here between drivers and regulator is itself a separate, longer-term story.

The second contrarian angle matters more. Safety sits at the centre of this incident — cars nearly stopping on a wet formation lap means the creation of accident risk. That risk is real and of the highest priority. But alongside it is another, less discussed risk: centralisation of decision-making. If a software layer governing energy deployment is centrally changeable, then how uniform each team's configuration is, and who verifies that uniformity, remains unanswered.

The third point is the limit of measurement. Reaching long-term conclusions from a single wet race weekend is risky. Driver dissatisfaction runs through the season, which increases sample size. But the sample of technical failure is still one. Merging the two means weighing two different kinds of information on the same scale.

— Root: FIA technical regulations / power-unit software layer | Scenario: governance deep dive.

Toward a conclusion: what is still outside the ledger

The FIA has pledged a joint review with teams and power-unit manufacturers. The answers that emerge from that review will set the next step: is the correction in the affected sectors permanent or temporary? Will wet-condition pre-season testing protocols become mandatory? And will the formation-lap procedure itself change?

In my assessment, the biggest risk right now is not at Sepang — it is at some future wet race weekend. A fault undetectable in dry testing can return when the conditions return. And if cars then stop on track, the question raised will be about safety, not merely reputation.

— Root: 2026 empty-stadium natural experiment / methodology parallel | Scenario: cross-domain method note.

— Root: Data Monk archetype / INTJ patience | Scenario: methodology note.

The final question is simple, but the answer is hard: when the pace of a sport depends on a decision made in code, who owns that code — the factory that builds the machine, or the institution that writes the rules? The rain at Sepang has stopped, but that question is still standing on the track.